# US 2021/0279850 A1 · System and Method for Identifying a Machine Tool Having Processed a Wood Piece

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0279850 A1
- Document ID: US20210279850A1
- Status: Published application
- Kind: A1
- Application number: 17/196,730
- Filed: 2021-03-09
- Published: 2021-09-09
- Earliest claimed: 2020-03-09
- Assignees: Bid Group Technologies Ltd
- Applicants: Bid Group Technologies Ltd.
- Inventors: Gabriel Beaudet; Francis Clement; Alexandre Prevost; Guy Morissette
- Web page: https://patentgrep.com/patent/US20210279850A1

## CPC

- G06V 10/44: Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06F 18/2431: Multiple classes
- G06T 7/001: {using an image reference approach}
- G06T 7/11: Region-based segmentation
- G06T 7/40: Analysis of texture (depth or shape recovery from texture G06T7/529)
- G06T 7/73: using feature-based methods
- G06V 10/764: using classification, e.g. of video objects
- G06T 2207/20076: Probabilistic image processing
- G06T 2207/20081: Training; Learning
- G06T 2207/30161: Wood; Lumber
- G06V 2201/06: Recognition of objects for industrial automation
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A system and method for detecting the origin of wooden planks in a sawmill is provided. The method scans surfaces of processed planks and, with the help of an AI algorithm comprising a deep-learning algorithm, determines the origin of said planks based on analysed parameters of the planks. The parameters used in the analysis are mainly properties of tool marks and the resulting analysis provides tools and equipment used. The deep learning algorithm may be in a self-learning mode or in a supervised training mode.

## Claims

### Claim 1

1) A system for identifying a machine tool having processed a wood piece comprising tool marks in a production line, the system comprising:
a computerized device executing a program; and
an information capturing device in communication with the computerized device, the information gathering device having access to wood pieces in the production line and being configured to:
capture quantifiable and/or qualitative characteristics of the wood pieces of the production line; and
communicate the said captured characteristics to the computerized device;
the program being configured to:
detect marks or patterns on a surface of the wood piece based on the received quantifiable and/or qualitative information; and
identify a machine tool of the production line having processed the wood piece based on the captured quantifiable and/or qualitative characteristics.

### Claim 2 (depends on 1)

2) The system of claim 1, the program being trained to detect marks or patterns on a wood piece.

### Claim 3 (depends on 2)

3) The system of claim 2, the program being trained using an artificial intelligence algorithm.

### Claim 4 (depends on 3)

4) The system of claim 3, the artificial intelligence algorithm using unsupervised training.

### Claim 5 (depends on 3)

5) The system of claim 3, the artificial intelligence algorithm using supervised training having predetermined categories of characteristics to analyse.

### Claim 6 (depends on 1)

6) The system of claim 1, the information gathering device being selected in one of the followings: a laser scanner, a camera-based scanner, a thermal scanner, and a sound wave-based scanner.

### Claim 7 (depends on 1)

7) The system of claim 1, the identification of the machine tool of the production line having processed the wood piece being based on a single capture of the quantifiable and/or qualitative characteristics of the wood pieces.

### Claim 8 (depends on 1)

8) The system of claim 1, the program being further configured to identify one or more irregularities of the detected marks or patterns, the irregularities being quantifiable and/or qualitative values of characteristics within a predetermined range of values.

### Claim 9

9) A computer-implemented method for identifying a machine tool or equipment having processed a wood piece in a production line, the method comprising:
scanning at least one surface of the wood piece;
processing each of the scanned surface to identify marks, signs and/or traits present on the at least one scanned face of the wooden piece;
analyzing characteristics of the processed marks, signs and/or traits; and
mapping the machine tool or equipment used to process the marks, signs and/or traits based on the analyzed characteristics.

### Claim 10 (depends on 9)

10) The method of claim 9 comprising scanning the at least one surface comprising capturing one or more images and/or geometrical data of one or more faces of the wooden piece.

### Claim 11 (depends on 10)

11) The method of claim 10, the method further comprising categorizing the marks, signs or traits in predetermined categories, the mapping of the machine tool and equipment being based on the categorized marks, signs, or traits.

### Claim 12 (depends on 11)

12) The method of claim 11, the mapping of the machine tool and equipment further being based on the categorized marks, signs or traits being located on the right, left, top or bottom of a central point.

### Claim 13 (depends on 11)

13) The method of claim 11, the method further comprising training a trained artificial intelligence algorithm to categorize the processed marks based on the characteristics of the processed marks.

### Claim 14 (depends on 13)

14) The method of claim 13, the method further comprising training the artificial intelligence algorithm using wood pieced having predefined marks signs or traits.

### Claim 15 (depends on 14)

15) The method of claim 14, the training using an unsupervised mode.

### Claim 16 (depends on 14)

16) The method of claim 14, the training using a supervised mode.

### Claim 17 (depends on 16)

17) The method of claim 16, the supervised artificial mode comprising using an identification table comprising machine tool or equipment information and associated predetermined marks, signs, or traits.

### Claim 18 (depends on 17)

18) The method of claim 17, the method further comprising updating the identification table during the training of the artificial intelligence algorithm.

### Claim 19 (depends on 9)

19) The method of claim 9, the method further comprising:
calculating a matching probability between the analyzed characteristics and a list of machine tools or equipment present in the production line;
comparing the calculated matching probability to a predetermined level; and
mapping a machine tool or equipment to the wood piece when the calculated matching probability is equal or higher than the predetermined level.

### Claim 20 (depends on 9)

20) The method of claim 9, the scanning of the surface capturing any one of the followings: geometrical data, color data, scalar data, information about the edges of the wood piece and thermographic data.

### Claim 21 (depends on 9)

21) The method of claim 9, the analyzed characteristics comprising any one of the followings: level of curvature, width, depth and length of the mark, sign or trait, distance between different marks, signs or traits and variation of the said identified characteristics.

### Claim 22 (depends on 9)

22) The method of claim 9, wherein the processed marks, signs or traits originate from any one of the followings: a band saw, a conic canter pattern, a cylindrical canter pattern, a circular saw or any combination thereof.

### Claim 23 (depends on 9)

23) The method of claim 9, the method comprising scanning two different wood pieces and matching the two different wood pieces as having been processed by a common machine tool or equipment of the production line.